DSM strategies to reduce electricity costs
on platinum mines
JA Deysel
21690014
Dissertation submitted in partial fulfilment of the requirements
for the degree
Magister
in Mechanical Engineering at the
Potchefstroom Campus of the North-West University
Supervisor:
Prof M Kleingeld
Abstract
Title: DSM strategies to reduce electricity costs on platinum mines Author: J.A. Deysel
Promoter: Prof M. Kleingeld
Degree: Master of Engineering (Mechanical)
Labour disruptions, operational cost increases, mineral price decreases and nationalisation are only a few of the financial risks faced by mining companies in South Africa. These risks are not only impacting the mining companies but have a rippling effect on the surrounding communities and the country’s economy.
With the South African Platinum Group Metals (PGM) largely affected, platinum mining companies are reviewing all possible cost savings strategies. Together with the help of Eskom’s Integrated Demand Management (IDM) programmes, various electricity cost savings strategies are being investigated. To date implemented DSM strategies have deemed themselves successful on numerous mines.
These DSM strategies range from replacing basic equipment to maintenance schedules. However, there is a need to evaluate DSM strategies specifically applicable to standard underground platinum mines. Therefore this study reviews several plausible DSM strategies for the largest service systems including compressed air, refrigeration, ventilation, water reticulation and dewatering systems on platinum mines.
The first step in the investigation evaluated each DSM strategy according to existing system operations and infrastructure. The second step determined the feasibility of each DSM strategies for specific platinum mine case studies. This was done by simulating the electrical energy impact and investigating the required infrastructure for each strategy. With the predicted results and required infrastructure, an estimated payback period and a five year financial prospect was calculated to determine the feasibility.
Three strategies were selected and successfully implemented on the platinum mine case study. With the delivered electrical energy impact the annual electricity cost savings was calculated at an average R14.5 million with the implementation of the three DSM electrical cost saving strategies. The study provides insight into selecting the best proven electricity cost savings strategy according to the layout and design of a platinum mine.
Acknowledgments
I would like to give thanks to the Lord Jesus Christ for blessing me with all my abilities to have completed my studies and most importantly for His unfailing love.
Secondly, I want to thank TEMM International (Pty) Ltd and HVAC International (Pty) Ltd for the opportunity, financial assistance and support to complete this study.
To my parents and other family members, thank you for everything that you have done for me. Without you I would not have had the opportunity to complete my studies.
Prof. M. Kleingeld, my promoter, thank you for all your input and suggestions.
A special thanks to Dr. L. van der Zee, my mentor, for all your guidance and help with reviewing my dissertation.
Miss. J. Dumas, thank you for proofreading my dissertation. To Miss. L. Visagie, thank you for all your love and motivation.
Lastly, I would like to thank all my friends and colleagues for their support and friendship throughout the last two years.
Table of Contents
Abstract ... ii Acknowledgments ... iii List of figures ... vi List of tables ... ix Abbreviations ... x Nomenclatures ... xi Chapter 1: Introduction ... 21.1 The South African platinum mining industry ... 2
1.2 Reducing electricity costs through DSM strategies ... 3
1.3 DSM potential on platinum mines ... 5
1.4 Objectives of this study ... 6
1.5 Dissertation overview ... 6
Chapter 2: Investigation of possible DSM strategies ... 9
2.1 Introduction ... 9
2.2 Compressed air systems ... 9
2.3 Refrigeration and ventilation systems ... 20
2.4 Water reticulation systems ... 31
2.5 Conclusion ... 38
Chapter 3: Evaluation of possible DSM strategies ... 40
3.1 Introduction ... 40
3.2 Compressed air systems ... 40
3.3 Refrigeration and ventilation systems ... 57
3.4 Water reticulation systems ... 73
3.5 Conclusion ... 81
Chapter 4: Implementation and verification of DSM strategies ... 83
4.3 Refrigeration and ventilation systems ... 92
4.4 Conclusion ... 101
Chapter 5: Conclusion and recommendations ... 104
5.1 Conclusion ... 104 5.2 Recommendations ... 105 Chapter 6: References ... 108 Appendix A ... 115 Appendix B ... 116 Appendix C ... 117 Appendix D ... 123 Appendix E ... 126
List of figures
Figure 1: 2012 PGM mining economical contribution. ... 2
Figure 2: Costs and expenditures of the platinum sector (Baxter, 2014). ... 3
Figure 3: Electrical energy distribution of the mining sector (Le Roux, 2005). ... 4
Figure 4: Average weekday Eskom Megaflex electricity tariff. ... 5
Figure 5: Typical layout of stand-alone compressed air network. ... 10
Figure 6: Typical ring-feed compressed air network. ... 11
Figure 7: Inefficient compressor usage at Mine B. ... 12
Figure 8: Compressed air leak wastage. ... 14
Figure 9: Poorly repaired air leak at platinum Mine A. ... 15
Figure 10: Rock drill power consumption (Willis, 2008). ... 15
Figure 11: Surface control valve utilised on Mine A. ... 17
Figure 12: Photo of a Brown Boveri Sulzer centrifugal compressor guide vanes. ... 18
Figure 13: Modulating- and Moore-control implementation on a centrifugal compressor. ... 19
Figure 14: Typical mine refrigeration and water reticulation system layout. ... 21
Figure 15: Vapour-compression cycle of a shell-and tube-heat exchanger. ... 22
Figure 16: Ammonia absorption cycle. ... 24
Figure 17: Typical cooling tower and vertical bulk air cooler (Amsted Industries, 2011; McPherson, 1993). ... 26
Figure 18: Load management power profile for Mine A. ... 27
Figure 19: Flow control power profile for Mine A. ... 28
Figure 20: Potential electrical energy savings and electrical motor speed. ... 29
Figure 21: Typical mine water reticulation system. ... 32
Figure 22: Typical GNM cooling car. ... 33
Figure 23: Water leak wastage. ... 34
Figure 24: Pressure control implemented on Mine C. ... 35
Figure 25: Multiple pumps performance curve. ... 36
Figure 26: Parallel and series pump arrangements. ... 36
Figure 27: Theoretical 1 MW load management profile (van der Zee, 2013). ... 37
Figure 28: Temporary power and air flow loggers. ... 41
Figure 29: Mine A compressed air network layout. ... 43
Figure 30: Mine A compressor power profile and operation schedule. ... 44
Figure 31: Mine A compressed air pressure requirements. ... 45
Figure 32: Compressed air system simulation of Mine A. ... 46
Figure 35: Mine A compressed air network and proposed infrastructure. ... 49
Figure 36: PID control example for Mine A (Krass, 2006). ... 50
Figure 37: Simulated pressure with downstream control on Mine A. ... 50
Figure 38: Mine A downstream control simulated power profile. ... 51
Figure 39: Moore compressor controller (Matrikon Moore, 2010). ... 52
Figure 40: Mine A compressor control simulated delivery pressure. ... 53
Figure 41: Mine A compressor control simulated power profile. ... 53
Figure 42: Mine A integration simulation results. ... 55
Figure 43: Estimated payback period for compressed air optimisation on Mine A. ... 56
Figure 44: Mine A refrigeration and ventilation layout. ... 59
Figure 45: Mine A refrigeration and ventilation simulated summer power profile. ... 64
Figure 46: Mine A refrigeration and ventilation simulated winter power profile. ... 64
Figure 47: Mine A refrigeration and ventilation system and proposed infrastructure. ... 66
Figure 48: Mine A refrigeration and ventilation simulated load management summer power profile. 67 Figure 49: Mine A refrigeration and ventilation simulated load management winter power profile. .. 68
Figure 50: Mine A refrigeration and ventilation simulated BAC summer wet-bulb temperature. ... 68
Figure 51: Mine A refrigeration and ventilation simulated variable water flow control summer power profile. ... 70
Figure 52: Mine A refrigeration and ventilation simulated variable water flow control winter power profile. ... 71
Figure 53: Estimated payback period for refrigeration and ventilation optimisation on Mine A. ... 72
Figure 54: Temporary water flow logger. ... 74
Figure 55: Mine A water reticulation system. ... 75
Figure 56: Mine A dewatering system. ... 76
Figure 57: Mine A water reticulation power profile. ... 77
Figure 58: Mine A 21 level collection dam level vs. inlet and outlet flow rates. ... 78
Figure 59: Mine A 21 level collection proposed dam level vs. proposed inlet and outlet flow rates ... 79
Figure 60: Downstream control valves installed on Mine A compressed air system. ... 86
Figure 61: Air mass flow meters and pressure transmitters installed on Mine A compressed air system. ... 86
Figure 62: PLC unit installed on Mine A compressed air system. ... 87
Figure 63: Pneumatic actuated inlet guide vane controller of Compressor 1 at Mine A. ... 87
Figure 64: Downstream pressure control of Shaft 3 compressed air control valve on Mine A. ... 88
Figure 65: Compressor 4 delivery pressure control on Mine A compressed air system. ... 88 Figure 66: Compressor 4 power consumption vs. delivery pressure on the compressed air system of
Figure 68: Mine A compressor electrical energy consumption before and after integrated strategy.... 91
Figure 69: Pneumatic actuating control valve installed on the condenser water cycle of Mine A. ... 92
Figure 70: 200 kW and 250 kW VSDs installed on the condenser and evaporator water pumps of Mine A. ... 93
Figure 71: 400kW VSDs installed on the condenser and evaporator water pumps of Mine A. ... 94
Figure 72: Average electrical energy profiles for the refrigeration and ventilation system of Mine A.96 Figure 73: Mine A refrigeration and ventilation electrical energy consumption before and after load management strategy. ... 97
Figure 74: Mine A post-implementation variable evaporator water flow control. ... 99
Figure 75: Mine A post-implementation daily water flow to shaft and cold dam level... 99
Figure 76: Figure 73: Mine A refrigeration and ventilation electrical energy consumption before and after variable water flow strategy. ... 100
Figure 77: Mine A post-implementation payback period results. ... 101
Figure 78: Appendix A; 2014/2015 Eskom Megaflex tariff rates. ... 115
Figure 79: Appendix B; Mine A refrigeration and ventilation system simulation model. ... 116
Figure 80: Appendix E; Mine A manual condenser water flow test 1. ... 126
List of tables
Table 1: The compressed air requirements of typical mining activities. ... 9
Table 2: Specifications of the refrigeration and ventilation system on Mine C (Du Plessis, 2013). .... 29
Table 3: Mine A compressor parameters. ... 43
Table 4: Simulated results for downstream control on Mine A. ... 51
Table 5: Simulated results for compressor control on Mine A. ... 54
Table 6: Mine A control set-points. ... 54
Table 7: Simulated results for downstream and compressor control on Mine A. ... 55
Table 8: Mine A refrigeration and ventilation system parameters. ... 61
Table 9: Mine A refrigeration and ventilation system operation schedules and parameters. ... 63
Table 10: Simulated results for a load management strategy on Mine A. ... 69
Table 11: Mine A refrigeration and ventilation optimised winter system operation schedules. ... 70
Table 12: Simulated results for variable water flow control on Mine A... 71
Table 13: Integrated control strategy bill of quantities for Mine A compressed air system. ... 84
Table 14: Compressed air profile scaling coefficient parameters proposed by M&V team. ... 90
Table 15: Mine A compressed air system savings summary. ... 91
Table 16: Load management and variable water flow strategy bill of quantities for Mine A refrigeration and ventilation system. ... 94
Table 17: Mine A load management and variable water flow control parameters. ... 95
Table 18: Mine A refrigeration and ventilation system savings summary. ... 100
Table 19: Mine A post-implementation electricity cost savings results. ... 102
Table 20: High electricity consuming platinum mines. ... 104
Table 21: Appendix C; Mine A downstream and compressor control quotations. ... 117
Table 22: Appendix C; Mine A load management and variable water flow quotation... 119
Table 23: Appendix D; Mine A integrated downstream and compressor control strategy post-implementation results. ... 123
Table 24: Appendix D; Mine A load management strategy post-implementation results. ... 124
Abbreviations
BAC Bulk Air Cooler
COP Coefficient of Performance CTM Critical Thermal Maximum DSM Demand Side Management
EEDSM Energy Efficiency Demand Side Management ESCO Energy Service Company
GDP Gross Domestic Product
IDM Integrated Demand Management M&V Measurement and Verification PGM Platinum Group Metals PI Proportional Integral
PID Proportional Integral Derivative PLC Programmable Logic Controller
PTB Process Toolbox
SCADA Supervisory Control and Data Acquisition VRT Virgin Rock Temperature
Nomenclatures
General: °C Degree Celsius g Gram h Hour J Joule k Kilo K Kelvin ℓ Litre m Meter M Mega Pa Pascal R Rand s Second t Tonne W Watt % Percentage Chapter 2: 𝐶 Leak coefficient𝐶𝑝 Specific heat constant
𝑚̇ Air mass flow rate
𝑛 Polytropic compression exponent
𝑃𝑒𝑙𝑒𝑐𝑡𝑟𝑖𝑐𝑎𝑙 Electrical power
𝑄̇ Heat transfer rate
𝑅 Gas constant 𝑇 Temperature 𝑤𝑐𝑜𝑚𝑝 Compressor energy Chapter 3: 𝐷𝑣𝑜𝑙𝑢𝑚𝑒 Dam volume 𝐿 Dam level
𝑚̇ Water mass flow rate
𝑡 Time
Chapter 4:
Integrated downstream and compressed control:
𝑃𝑒𝑙𝑒𝑐𝑡𝑟𝑖𝑐𝑎𝑙 Electrical power
𝑝 Compressor delivery pressure
𝑣 Compressor delivery volumetric flow
∅ Average compressor delivery pressure coefficient
𝜑 Average compressor delivery volumetric flow rate coefficient 𝛼 Compressed air interception coefficient
Load management:
𝑃𝑎𝑐𝑡𝑢𝑎𝑙 Actual electrical power 𝑃𝑛𝑜𝑟𝑚𝑎𝑙 Normal electrical power
𝛼 Refrigeration and ventilation electrical profile scaling factor
Variable water flow control:
𝑃𝑠𝑐𝑎𝑙𝑒𝑑 Scaled electrical power
𝑇ℎ𝑜𝑡 𝑑𝑎𝑚 Hot dam temperature
𝑇𝑐𝑜𝑙𝑑 𝑑𝑎𝑚 Cold dam temperature
𝑣𝑠ℎ𝑎𝑓𝑡 Water flow rate to shaft
𝑣𝐵𝐴𝐶 Water flow rate to BAC
Chapter 1
Introduction
Chapter 1: Introduction
1.1 The South African platinum mining industry
To date, South Africa is one of the leading mineral producing countries in the world, with a mineral reserve worth R20.3 trillion. Of these reserves 66.3% of South Africa’s 2011 mineral exports came from Platinum Group Metals (PGM), coal and gold (Kearney, 2012).During 2011 the mining sector also contributed 8.8% of the direct Gross Domestic Product (GDP) (Yager et al., 2013). And in 2012, the PGM mining sector alone accounted for 4.1% of the country’s GDP and 9% of its merchandise exports (Baxter, 2014). Figure 1 illustrates the PGM contribution to the GDP.
Figure 1: 2012 PGM mining economical contribution.
Platinum is currently considered as one of the scarcest metals in the world, with just over 200 metric tons being produced globally each year (Yang, 2009), with statistics showing that 72% of the world’s platinum produce originated from South Africa (Yager et al., 2013). However, the South African PGM production is annually declining by 0.4% due to lower global demand, increased domestic costs and labour disruptions (Baxter, 2014).
In 2014 the platinum producing companies in South Africa lost approximately R24 billion in revenues due to a five month labour disruption which started on 23 January 2014 (Platinum producers, 2014). Anglo American Platinum released a report claiming a 5.24 metric ton platinum loss within the first quarter alone (Anglo American, 2014). The labour actions are not only affecting production but also the platinum price which has seen a 12% decrease since 2011 (Magara, 2014).
To mitigate the losses, platinum mining companies are reviewing possible restructuring and lowering of costs and expenditures. However, if the financial deficit does not improve, extreme actions such as
0% 2% 4% 6% 8% 10% 12% 14% 2008 2009 2010 2011 2012
cannot be lowered without large scale restructuring. But the fourth largest contributor, electricity, can be further reviewed for possible cost reductions. A breakdown of the costs and expenditures of the platinum sector is displayed in Figure 2 (Baxter, 2014).
Figure 2: Costs and expenditures of the platinum sector (Baxter, 2014).
1.2 Reducing electricity costs through DSM strategies
Electricity is one of the most important commodities in the modern world and is slowly but surely becoming one of the most expensive commodities in South Africa. In 2013 Eskom introduced their new Multi-Year Price Determination (MYPD 3) plan three which includes an annual 8% tariff increase from 2013 to 2018.With this tariff increase, the current (2014/15) average electricity price of 70.75 c/kWh will increase to 89.13 c/kWh by 2018 (Creamer, 2013).
With the mining sectors consuming 14.1% of the total Eskom electricity sales, the tariff increase will only worsen the fragile financial state of mining companies and in turn the South African economy (Eskom Ltd., 2014). Therefore, platinum mining companies, such as Anglo American, together with Eskom have embarked on electrical energy savings projects to reduce the demand-side load (Anglo American, 2014). These electrical energy savings projects, also called Demand Side Management (DSM) Projects, form part of the Integrated Demand Management (IDM) Programme (Eskom Ltd., 2012).
DSM projects are subdivided into four categories namely Energy Efficiency DSM (EEDSM), Demand Response (DR), Energy Conservation Scheme (ECS) and Electricity Growth Management (EGM) (Eskom Ltd., 2014). The main focus of Energy Service Companies (ESCOs) and platinum mining companies are to implement EEDSM projects on high electrical energy consuming activities.
Labour costs (C band and above) 28% Stores and materials 16% Long-term project capex 12% Stay-in-business capex 9% Labour costs (D band and above) 8% Electricity costs 6% Steel costs 3% Other costs 28%
The typical electrical energy distribution of the underground mining sector is depicted in Figure 3 (Le Roux, 2005).
Figure 3: Electrical energy distribution of the mining sector (Le Roux, 2005).
The platinum mining operations can be divided into two categories; production and services (Schutte, 2013). Production includes all the activities such as drilling, blasting, sweepings etc. involved in mining the platinum ore. Whereas services include all the systems, i.e. compressed air, refrigeration, ventilation, hoisting and pumping that are utilised by the production activities. These five service systems are essential for any underground mine but consumes 61.1% of all the electrical energy supplied to the mining sector.
With the service systems consuming the majority of electrical energy, it presents the opportunity to investigate and possibly implement three different DSM interventions namely electrical energy efficiency, load management and peak clip strategies. Electrical energy efficiency strategies will reduce the load throughout the day according to the user demand. However, load management and peak clip will focus on reducing the load during periods of the day when the electricity tariffs are the highest.
Each of the DSM strategies will be implemented to accommodate the production schedules and the Eskom tariff structures. Figure 4 incorporates both the mining production schedules and the weekday 2014/2015 Eskom Megaflex electricity tariff plan (Eskom Ltd., 2014). The average electricity tariff for the high demand season (June - August) and the low demand season (September - May) was used in Figure 4 and will be used throughout the dissertation. The Eskom Megaflex electricity tariff plan for 2014/2015 is shown in Appendix A.
Compressors 21.3% Mining activities 18.9% Pumping 17.7% Hoisting 14.2% Mineral processing plants 13.7% Refrigeration and ventilation 7.9% Office buildings, hostels and services 8%
Figure 4: Average weekday Eskom Megaflex electricity tariff.
1.3 DSM potential on platinum mines
Until recently the price of electricity in South Africa was among the lowest in the world (NUS Consulting Group, 2007). This presented mining companies the opportunity to operate oversized auxiliary equipment to achieve maximum production without any concerns regarding electrical energy efficiency. With over 46% of the mining sectors electricity consumption being allocated to these systems, the potential for savings is substantial (Le Roux, 2005).
1.3.1 Compressed air systems
Compressed air is perhaps the most used and most expensive utility on any mine (Botha, 2011). Mining companies constantly overdesign compressed air systems to avoid any shortcomings due to the different requirements needed for each user. Not only are these systems overdesigned, but they are operated inefficiently due to a lack of supply and demand control. However, various DSM strategies implemented on compressed air systems have reported ample electrical energy savings.
These strategies have incorporated compressor control, downstream control and regular maintenance to achieve the savings (Schutte, 2013). This dissertation will investigate the feasibility of potential compressed air DSM strategies and implement them within the platinum mining sector.
1.3.2 Refrigeration and ventilation systems
With most South African platinum mines being situated within the Bushveld Igneous Complex, the Virgin Rock Temperatures (VRT) can increase up to 18°C per kilometre in depth (Venter, 2007). To overcome these temperatures and create a suitable underground working environment, platinum mines
0 20 40 60 80 100 120 140 160 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 E lect ricit y t a rif f [c/k Wh] Time [hour]
Average annual electricity tariff [c/kWh]
Off peak Peak Standard
Standard Peak
Off peak
utilise large refrigeration and ventilation systems. However, these systems are constantly operated at a maximum load throughout the day and year.
This presents the opportunity to investigate possible DSM strategies to reduce the electricity costs of these systems. The DSM strategies will incorporate load control and supply control in accordance with the mine health and safety regulations.
1.3.3 Water reticulation systems
Water is one of the essential utilities needed for underground mining. It has a wide variety of uses ranging from secondary cooling to suppression of dust (Schutte, 2013). However, any water within a mine, fissure as well as rain water, will accumulate at a central location at the bottom of the shaft. For dewatering, mining companies utilise large centrifugal pumps to move water from a central location, usually dams, to another until it reaches the surface (de la Vergne, 2003). The dewatering pumps, dams and piping networks all form the water reticulation system.
These water reticulation systems are usually manually operated without any specific control. This again presents the opportunity to investigate possible DSM strategies which will incorporate the automating of the entire water reticulation system. The automation will include proper control strategies to accommodate the Eskom Megaflex tariffs and mining schedules.
1.4 Objectives of this study
The current platinum crisis presents a vast need for system restructuring and optimisation interventions to mitigate the already fragile state. The objective of this study will be to investigate existing DSM strategies and adapt them specifically to underground platinum mines. The implementation of these DSM strategies will not only reduce the electricity costs but lower the operational overheads.
1.5 Dissertation overview
The overview of this dissertation is discussed below: Chapter 1
The introduction provides background to the current South African platinum mining crisis and the intervention required to reduce operational costs. It outlines the typical underground mine electrical energy distribution and existing DSM strategies. The outcome of this chapter is to identify possible DSM strategies on compressed air, refrigeration, ventilation and water reticulation systems.
Chapter 2
In this chapter an overview of the typical compressed air, refrigeration, ventilation and water reticulation systems are presented and discussed in depth. Thereafter, potential electricity cost saving strategies through system improvements or alterations are investigated. The objective of this chapter is to provide a basis for the simulations and optimisation models in Chapter 3.
Chapter 3
The purpose of this chapter is to investigate the feasibility of implementing DSM strategies on several platinum mine case studies. These strategies are then simulated and optimised with the use of a thermal hydraulic simulation flow solver.
Chapter 4
This chapter provides insight into the implementation and integration of the DSM strategies. Hereafter, the results achieved from each case study is verified against the proposed results from Chapter 3.
Chapter 5
To conclude the dissertation, a review of the problem statement, processes followed to reach a valid solution and the findings are summarised in this chapter. Finally recommendations for future studies on electricity cost saving strategies for platinum mines are discussed.
Chapter 2
Investigation of possible DSM strategies
Classification of DSM strategies applicable to the platinum mining industry and its high electricity consuming service system
Chapter 2: Investigation of possible DSM strategies
2.1 Introduction
Chapter 1 briefly discussed the potential for electricity cost saving strategies on the significant electrical energy consuming systems of South African underground platinum mines. Chapter 2 investigates and provides background on the compressed air, refrigeration, ventilation and water reticulation systems of a typical underground platinum mine. Attention will then be given to possible electricity cost savings through the implementation of existing system alterations and/or improvements.
2.2 Compressed air systems
2.2.1 Compressed air usage throughout the mining industry
A compressed air system is of high importance to most underground South African mines. Compressed air is used throughout the mine from the drilling process all the way to the processing plant. Each of these processes has its own specific requirements in terms of flow, pressure and quality of air. Table 1 outlines the compressed air requirements for a variety of mining activities (Brake & Bates, 1999; Joubert, 2010; Snyman, 2011).
Table 1: The compressed air requirements of typical mining activities.
Process Pressure [kPa] Pressure Category Flow
[m³/h] Flow Category
Agitation 400 Medium 1500 High
Drilling 450-600 High 160-1500 High
Workshops 250 Low 500 Medium
Hoisting 350 Medium 12 Low
Loading 350-860 High 93.6-1080 Medium
Refuge bays 200-300 Low 36 -300 Low
Cleaning 250 Low 2000 High
Tramming 350 Medium 12 Low
Pneumatic valves 350 Medium 12 Low
Plant equipment
and instrumentation 450 High 288-2520 Medium-High
<300 kPa = Low pressure >350 kPa & <450 kPa = Medium pressure >450 kPa = High pressure <500 m³/h = Low flow >500 m³/h & <1500 m³/h = Medium flow >1500 m³/h = High flow Large multistage compressors are being used by South African mines to deliver the required amount of compressed air. These compressors deliver compressed air at a flow rate of 10 000 – 60 000 m³/h and a pressure of 500 – 600 kPa (Snyman, 2011).
A compressed air network can comprise of kilometres of piping on surface and underground to distribute the compressed air. Over these long distances the air will never reach the end user at the same (flow and pressure) condition as it was originally supplied by the compressors. This is due to a variety of factors such as auto compression, pipe friction losses, air leakages and unauthorised users. Some of these factors can be minimised with regular inspection and maintenance of the airline. A typical air network will have multiple compressor houses with one or more compressors supplying air. These compressors will most likely vary in size and capacity and will be situated in compressor houses close to a compressed air user such as a shaft or a processing plant. Compressed air networks can be categorised as either a stand-alone system or a ring-feed system as illustrated in Figure 5 and Figure 6 (Joubert, 2010).
Figure 5: Typical layout of stand-alone compressed air network.
Stand-alone systems consist of a single airline from the supplier to the user. Whereas ring-feed systems utilise a closed ring airline where all the suppliers feed into and the users feed out of. Both these systems have their advantages and disadvantages but a ring-feed system is the most preferred. This is due to the fact that the compressed air supplies are spread throughout the network and this facilitates maintenance schedules without a complete shutdown.
All compressed air users can be grouped into categories of pressure and flow as shown in Table 1 (Schutte & Kleingeld, 2010). With different users demanding compressed air from the same network, it is difficult to supply compressed air at the required flow rate and pressure. Therefore mines schedule the run time of their compressors to pressurise the compressed air network according to the requirements of its highest user (Marais, 2012; Venter, 2012) .The compressed air demand will also
Compressor house SCADA P F P F Compressor 1 Compressor 2
Shaft Processing plant or Concentrator
Level A Level C Level B Legend Compressor Control valve Manual valve Shaft Local SCADA Shaft walls Piping network Communication network Processing plant/ Concentrator Pressure transmitter Flow transmitter F P
pressure of 480 kPa is required on surface and underground. While during non-drilling shifts a minimum pressure of 350 kPa is required for agitation and refuge bays (Snyman, 2011).
Figure 6: Typical ring-feed compressed air network.
The run schedules for the different compressors are determined by the demand during the day. But irregular compressed air usage during the day can result in compressors being started and stopped not according to their schedules. South African underground mines use either operators or automated control systems to start and stop compressors according to the pressure demand. Both controlling methods can result in high compressor power consumption with the frequent starting and stopping of compressors during peak and off-peak periods (Marais, 2012). An example of an inefficient compressor usage at platinum Mine B is illustrated in Figure 7.
Level A Level C Level A Level B Level C Compressor house 2 SCADA Compressor 3 Compressor 4 Compressor house 1 SCADA Compressor 1 Compressor 2 Shaft 1 Processing plant Shaft 2 Concentrator P F F P P F F P Legend
Compressor Control valve
Manual valve Shaft Local SCADA Shaft walls Piping network Communication network Processing plant/ Concentrator Pressure transmitter Flow transmitter P F
Figure 7: Inefficient compressor usage at Mine B.
To fully understand the impact of unnecessary compressor usage the focus has to be shifted to the electrical power consumption. Compressors not only vary in size and capacity but also in the amount of electrical power required to generate compressed air. Equation 1 can be used to estimate the theoretical electrical power required to generate compressed air at atmospheric conditions (101.3 kPa, 20°C) (Cengel et al., 2011). Equation 1: 𝑃𝑒𝑙𝑒𝑐𝑡𝑟𝑖𝑐𝑎𝑙 =𝑚̇𝑎𝑖𝑟𝜂𝑤𝑐𝑜𝑚𝑝,𝑖𝑛 𝑚𝑜𝑡𝑜𝑟 Where: 𝑃𝑒𝑙𝑒𝑐𝑡𝑟𝑖𝑐𝑎𝑙 = electrical power (kW)
𝑚̇𝑎𝑖𝑟 = compressed air mass flow rate (kg/s)
𝑤𝑐𝑜𝑚𝑝,𝑖𝑛 = energy required to compress a unit mass of air (kJ/kg) 𝜂𝑚𝑜𝑡𝑜𝑟 = efficiency of the electrical motor (90%)
From Equation 1 it is evident that by decreasing the compressed air mass flow rate, the electrical power will decrease as they are directly proportional. Furthermore Equation 1 shows that the motor’s efficiency is inversely proportional to the electrical power. The electrical power is also dependent on the design of the compressor and the maintenance, which is incorporated into the electrical energy required to compress a unit mass of air (van der Zee, 2013). The electrical energy can be calculated using Equation 2 (Cengel et al., 2011).
0 4000 8000 12000 16000 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 P ow er [ kW] Time [hour]
Compressor 1 Compressor 2 Compressor 3 Compressor 4
Unnecessary stopping and starting of compressors.
Equation 2: 𝑤𝑐𝑜𝑚𝑝,𝑖𝑛= 𝑛𝑅𝑇𝑖𝑛 𝜂𝑐𝑜𝑚𝑝(𝑛 − 1)[( 𝑝𝑜𝑢𝑡 𝑝𝑖𝑛) (𝑛−1) 𝑛 ⁄ − 1] Where:
𝑤𝑐𝑜𝑚𝑝,𝑖𝑛= energy required to compress a unit mass of air (kJ/kg)
𝑛 = polytropic compression exponent 𝑅 = gas constant (0.287kJ/kg.K) 𝑇𝑖𝑛 = inlet temperature (K)
𝜂𝑐𝑜𝑚𝑝 = compressor efficiency
𝑝𝑜𝑢𝑡 = compressor discharge pressure (kPa)
𝑝𝑖𝑛 = compressor inlet pressure (kPa)
2.2.2 Compressed air system alterations and improvements
Studies have shown that between 20% and 50% of electrical energy savings can be realised with improvements on a compressed air system. Therefore, the following section will investigate and discuss compressed air improvement opportunities that can result in possible electricity cost savings (Hongbo & McKane, 2008; Joubert, 2010).
According to the first law of thermodynamics, the energy in a system is conserved and cannot be created or destroyed. The amount of energy used to compress air into a system will be equal to the energy used and lost by the system. Therefore, this section will first focus on reducing the amount of energy demand and losses in a compressed air system.
Air leaks
Leaks in a compressed air system can be a nuisance to most mine personnel, but leaks can be accountable for up to 20% of its total compressed air capacity (Hongbo & McKane, 2008). An air leak with a diameter of 10 mm might not seem large, but if unfixed it will have cost the mine approximately R86 000.00 per annum with the 2014 electricity rates (Eskom Ltd., 2014). Repairing an air leak is a small price to pay compared to the price of electricity. Figure 8 illustrates the amount of wasted electrical power and cost (Eskom 2014/2015 Megaflex tariffs) according to the diameter of the air leak (Eskom Ltd., 2014; van der Zee, 2013).
Figure 8: Compressed air leak wastage.
Figure 8 was calculated using both Equation 1 and Equation 2. The assumed parameters are: 𝑛 = polytropic compression exponent (1.4)
𝑘 = specific heat ratio for air (1.4) 𝑅 = gas constant (0.287 kJ/kg.K) 𝐶 = coefficient of leak (0.65) 𝑇𝑖𝑛 = inlet temperature (25°C) 𝑇𝑙𝑖𝑛𝑒 = line temperature (28°C)
𝑝𝑖𝑛 = compressor inlet pressure (87 kPa) 𝑝𝑙𝑖𝑛𝑒 = compressor discharge pressure (500 kPa) 𝜂𝑐𝑜𝑚𝑝 = compressor efficiency (80%)
𝜂𝑚𝑜𝑡𝑜𝑟 = motor efficiency (90%)
In addition to the wasted electrical power, air leaks can cause reduced tool efficiencies due to downstream pressure drops. Reduced air tool efficiencies will also lead to lower production, increased operating hours and additional maintenance (Hongbo & McKane, 2008). According to van der Zee it will be uneconomical to repair each small leakage due to the fact that the entire airline has to be
R 0 R 300 R 600 R 900 R 1 200 0 400 800 1200 1600 2 10 18 26 34 42 50 58 66 74 82 90 98 E lec tri ci ty cos t per hou r [R and ] E lec tri cal pow er [ kW] Leak diameter [mm]
2013). Figure 9 shows an example of a surface air leak that has been poorly repaired by mine personnel on platinum Mine A.
Figure 9: Poorly repaired air leak at platinum Mine A.
Pneumatic equipment
Another possibility is to replace current pneumatic equipment with either electrical or hydropowered equipment. Figure 10 illustrates the average power consumption of pneumatic rock drills compared to electrical and hydropowered rock drills (Willis, 2008).
Figure 10: Rock drill power consumption (Willis, 2008).
Unfortunately studies performed at Anglo American Platinum in 2008 have shown that the operational cost of electrical rock drills would be approximately 28% more than pneumatic rock drills (Petit, 2006). A similar study confirmed these results by showing that the operational costs for
0 2 4 6 8 10 12 14 Pneumatic drill (350 kPa) Pneumatic drill (500 kPa)
Hydropowered drill Electrical drill
C onsume d pow er [ kW]
electrical rock drills can be up to 2.6 times more than pneumatic rock drills (Labuschagne, 2010). Although the results appear positive in terms of electrical power consumption, replacing pneumatic equipment will be uneconomical in terms of installation and operational costs.
Downstream control
Deep-level mines are notorious for over-pressurised air networks. The main reason for this is that some air networks feed multiple shafts and processing plants. Table 1 only outlines the required compressed air pressure and flow for when a process is in progress. When these processes end the air demand will change, but the supplied compressed air remains unchanged.
To eliminate or limit the above mentioned scenario three main strategies have been developed, namely: underground and surface-isolation, control and supply side control. Underground and surface-isolation is done by installing open/close isolation valves upstream on the supply airlines. The isolation valves will be controlled according to shift changes. Once a drilling shift has ended the valve will close isolating the supply air. When the next shift, for instance a cleaning shift starts, the isolation valve will open supplying the line with compressed air.
There are various methods for isolating an airline but the most frequently used method is manual isolation by designated personnel. The preferred method is a controlled isolation valve which can be opened or closed from a central point such as a control room to ensure isolation according to the schedule. These isolation valves can either be manually controlled by an operator or automatically controlled. Both these methods can be performed through a local Supervisory Control and Data Acquisitioning (SCADA) system (Joubert, 2010).
However, not all shafts can be isolated as there are refuge bays and loading bins that still require a constant pressure and flow rate. In order to solve this problem the isolation valve can be exchanged for a pressure/flow controlling valve. These valves will control the compressed air flow according to the downstream demand. A control valve will utilise an actuating positioner to control the valve according to the downstream pressure and flow readings (Joubert, 2010). Figure 11 shows an isolation and control valve setup that was installed on platinum Mine A.
In this setup a 250 mm control valve was utilised for refined control instead of a 450 mm control valve. Also due to the fact that only the downstream pressure would be affected, pneumatic actuating control valves were used and not electric actuating control valves.
Figure 11: Surface control valve utilised on Mine A.
Compressor control
The last optimisation methods focus on the supply side of a compressed air network. A compressor is designed to deliver compressed air at a specific pressure and flow but it is not limited to this design. Air is sucked into the compressor through the inlet valves where it is guided into the compressor through a set of guide vanes. Hereafter the air is compressed by rotor blades to a pre-set pressure and discharged into the air network through a controlling discharge valve. Before the air is discharged into the system, the compressor has to reach a high enough pressure to overcome back pressure from the system.
Each compressor is designed to operate within a set of stable parameters to prevent the compressor from entering its surge or choke point. Surge occurs once the flow through the compressor is too low to overcome the flow through the compressed air system and will result in flow reversal. Whereas at the choke point the inlet flow through the compressor is too high and a low head is developed. During this period the flow through the compressor will reach Mach 1.0. Both these conditions will cause extreme vibration and might damage the compressor and components (Barletta & Golden, 2004; Booysen et al., 2010; US DOE-OIT, 2003).
In order to prevent surge, a compressor is equipped with adjustable inlet valves, guide vanes and a blow-off valve. Once surge occurs the blow-off valve will immediately open to increase the flow
the desired inlet flow while the discharge and blow-off valves are closed (Blanchini & Giannattasio, 2002; US DOE-OIT, 2003; Willems et al., 2002). Figure 12 illustrates the adjustable inlet guide vanes of a RI-565 Brown Boveri Sulzer 4 2876 m³/h centrifugal compressor.
Figure 12: Photo of a Brown Boveri Sulzer centrifugal compressor guide vanes.
There are two methods of utilising the adjustable guide vanes and blow-off valves to reduce the amount of electrical power drawn by a compressor. These methods are either by placing a compressor in an unloaded state or by controlling the inlet vanes to reduce the delivery pressure.
The load/unload control of a compressor, known as constant-speed control, can be used to isolate the compressor from the compressed air system. This is done by closing the discharge valves and opening the blow-off valves. The compressor will still operate at a constant speed but will only consume up to 35% of its full-load capacity (US DOE-OIT, 2003). Unfortunately this controlling method can also be seen as energy wastage due to the compressed air being blown into the atmosphere.
Modulating control is the operation where the inlet valves are closed to vary the inlet air flow according to a pressure set-point. A more efficient control can be achieved by adjusting the inlet guide vanes angles to direct the inlet air straight into the impeller inlet. To prevent a compressor from entering a surge, the inlet vanes are limited to a minimum of 40% and maximum of 100% modulation. This modulating control has a direct effect on the discharge pressure and the electrical power consumed by the electrical motor (US DOE-OIT, 2003).
By combining the modulating control on a compressor with a programmable pressure profile, also
Adjustable inlet guide vanes
Figure 13 shows the reduction in consumed power and inlet air flow by applying modulating control and a Moore controller on a GHH Borsig centrifugal compressor.
Figure 13: Modulating- and Moore-control implementation on a centrifugal compressor.
DSM projects on compressed air systems normally vary between electrical energy efficiency and a peak clip project. The type of DSM project depends on the compressed air demand and supply of a mine during the day. An electrical energy efficiency project will focus on optimising the compressed air system to reduce the total electrical energy consumption throughout the day. Whereas, a peak clip project will focus only on reducing the electrical energy consumption during the Eskom peak periods. Chapter 3 will compare each system improvement according to the cost of the required infrastructure, the implementation period and the possible electrical energy savings. Thereafter, the DSM project selection will be discussed.
0 7500 15000 22500 30000 37500 45000 0 750 1500 2250 3000 3750 4500 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 V ol ume tri c fl ow rat e [m ³/ h] P ow er [ kW] Time [hour]
2.3 Refrigeration and ventilation systems
This section will discuss the function and operation of platinum mine refrigeration and ventilation systems. A short description of a typical platinum mine water reticulation system will also be provided as it runs concurrent with the refrigeration system.
2.3.1 Refrigeration and ventilation usage throughout the mining industry
Everyone and everything on the earth has a set of limitations. These limitations range from the temperature it can withstand to the mass it can uphold. Beyond these limitations failure is bound to occur whether it is a machine or a human. Therefore, a safe set of boundaries has been developed over the years for each and everything including humans. For underground mine workers these boundaries are being pushed to the limits every day.
One of the most important human boundaries is the Critical Thermal Maximum (CTM), which is the maximum temperature the human core can withstand, this ranges from 41.6°C to 42°C. Various studies have shown that in hot humid working condition the human core can easily reach temperatures above 39°C. However, when the human core reaches temperatures close to the CTM, dehydration and heat stroke becomes a reality (Bynum et al., 1978; Kosaka et al., 2004; Wyndham et al., 1966). An increase in working temperatures is not only dangerous for the health and safety of mine workers but it is inversely proportional to their performance (Le Roux, 1990). With this in mind the mining industry of South Africa introduced a maximum wet-bulb working temperature of 27.5°C. Another aspect to consider when choosing underground working conditions is the limitations of mining equipment. These limitations have been tested and will be specified by the manufacturers (Vosloo et
al., 2012). High operating temperatures for mining equipment will result in increased maintenance
down time and operational costs.
High temperatures in underground mines are caused by a variety of heat sources. Some of these heat sources are wall- and blasted-rock, adiabatic compression, electromechanical equipment, explosives, as well as ground and service water (Nixon et al., 1992). Of these heat sources, the rock walls of a mine are the largest source of heat. The VRT, depending on the thermal conductivity, can increase up to 18°C for every kilometre down (Gosh & Patterson, 1940; Nixon et al., 1992).
It is, therefore that deep-level mines utilise large refrigeration and ventilation systems to optimise the underground working conditions. Figure 14 illustrates a simplified layout of a typical refrigeration and ventilation system that incorporates chilled water reticulation (Du Plessis, 2013; Schutte, 2007). Water reticulation systems will be discussed in detail in Chapter 2.4.
Figure 14: Typical mine refrigeration and water reticulation system layout.
Figure 14 is a generalised layout due to the fact that the cooling and chilled water requirements for each underground mine is unique. Refrigeration and water reticulation systems can be subdivided into closed- or semi-closed loop systems. Figure 14 represents a semi-closed loop system where chilled water is required underground for mining equipment, Bulk Air Coolers (BACs) and spot coolers. A closed loop system will only be utilised for ventilation purposes and not chilled service water (Du Plessis, 2013; Schutte, 2013).
To summarise a water reticulation system, chilled water is supplied to underground users at a surface temperature between 3°C and 6°C. After each end user, the water will flow to storage dams on various underground levels where it is pumped back to surface. Here the water is collected in a hot dam with temperatures ranging from 25°C to 35°C. The underground water requirements will vary according to the size of a mine and its operation during the day (Calitz, 2006; Du Plessis et al., 2013).
Chilled water users
Storage dam Shaft ventilation air
Chilled water users Pre-cool tower Bulk air cooler
Evaporator water pump
Water transfer pump
Water turbine
Water reticulation pumps
Condenser water pump
Storage dam
Condenser cooling tower
Hot dam
Chilled water dam Expansion valve Condenser Evaporator Compressor Legend Compressor Control valve Bulk air cooling tower Cooling tower Condenser/ Evaporator Fan Dam Evaporator cycle Condenser cycle Refrigeration cycle Air Pump
Refrigeration system layout
A refrigeration system consists of an evaporator, condenser and refrigeration cycle. Within the refrigeration cycle, also known as a chiller, thermal energy is transferred from the evaporator to the condenser by means of a refrigerant fluid as seen in Figure 15. The two common refrigerant fluids used in the mining industry are ammonia or R134a. It is important to consider the working conditions of the fluid and the equipment to be used when selecting the appropriate refrigerant fluid (Borgnakke & Sonntag, 2009).
Figure 15: Vapour-compression cycle of a shell-and tube-heat exchanger.
As illustrated in Figure 15, after the thermal heat is transferred to the refrigerant vapour, from the evaporator water it is forced to the condenser by an electrical driven compressor. Within the condenser the thermal heat is transferred from the refrigerant to the condenser water. The refrigerant will also change from its vapour phase to a liquid phase due to the reduction in heat content - hence the use of an expansion valve to restore the refrigerant to a wet vapour with the reduction in pressure. Equation 3 can be used to determine the rate at which heat is transferred within the evaporator (Borgnakke & Sonntag, 2009).
W
C Condenser Evaporator Compressor Expansion valve eQ
cQ
CompressorLegend Expansion valve Heat absorption Heat emission Evaporator cycle Condenser cycle Refrigeration cycle cQ
eQ
Equation 3:
𝑄̇𝑒= 𝑚̇𝑤𝑐𝑝,𝑤(𝑇𝑤,𝑜− 𝑇𝑤,𝑖)
Where:
𝑄̇𝑒 = rate of heat transfer (W) 𝑚̇𝑤 = mass flow rate (kg/s)
𝑐𝑝,𝑤 = specific heat of water at constant pressure (J/kg.K) 𝑇𝑤,𝑜 = temperature of water out (K)
𝑇𝑤,𝑖 = temperature of water in (K)
A similar equation can be used to determine the rate at which heat is absorbed in the condenser, refer to Equation 4.
Equation 4:
𝑄̇𝑐= 𝑚̇𝑤𝑐𝑝,𝑤(𝑇𝑤,𝑜− 𝑇𝑤,𝑖)
Where:
𝑄̇𝑐 = rate of heat absorbed (W) 𝑚̇𝑤 = mass flow rate (kg/s)
𝑐𝑝,𝑤 = specific heat of water at constant pressure (J/kg.K)
𝑇𝑤,𝑜 = temperature of water out (K)
𝑇𝑤,𝑖 = temperature of water in (K)
Theoretically the total heat transferred in the evaporator must be equal to the heat absorbed in the condenser. This is called an adiabatic process and can only be used for theoretical calculations. In reality this process cannot be assumed as adiabatic as heat loss will occur.
Figure 15 represents a vapour-compression cycle but an ammonia absorption cycle can also be used for mine refrigeration systems. The ammonia absorption cycle exchanges the compressor for a liquid pump or a screw compressor. The liquid pump is used to pressurise an ammonia water solution through the cycle to a condenser. With the ammonia being absorbed in water, the specific volume will be less than that of a vapour which results in less work input required. Unfortunately the absorption
cycle is not entirely economically justifiable due to the amount of equipment and maintenance required as shown in Figure 16 (Borgnakke & Sonntag, 2009).
The guide vanes and slide valves of the compressor in a refrigeration cycle can be used to control the cooling load. This is achieved by altering the flow rate of the refrigerant throughout the cycle and therefore controlling the amount of heat transfer. The compressor control is based on the difference between a set-point and the actual value of evaporator outlet water temperature. By increasing the refrigerant flow rate, the electrical power usage and the cooling load will increase. Thus the electrical power usage of the compressor is directly related to the cooling load.
Figure 16: Ammonia absorption cycle.
With the rate of heat transfer in the evaporator and input electrical power of the compressor available, the cooling efficiency of the refrigeration cycle can be determined through Equation 5. The cooling efficiency, also known as the coefficient of performance (COP), indicates the amount of thermal cooling achieved for a unit of electrical power drawn by the compressor. A high COP value indicates an efficient energy usage and a low COP value the opposite (Du Plessis, 2013). Equation 5 can be used to determine the COP value for a single refrigeration cycle, chiller, or an entire refrigeration plant. This is done by substituting the electrical power input of a compressor by that of the entire plant (Borgnakke & Sonntag, 2009; Schutte, 2013).
p
W
Condenser Evaporator Pump Expansion valve HQ
LQ
Absorber Heat exchanger Generator Weak ammonia solution Strong ammonia solution High-pressure ammonia vapor Liquid ammonia Low-pressure ammonia vapor LQ
HQ
From high-temperature source Legend Liquid pump Expansion valveHeat transfer rate Ammonia absorber Evaporator cycle Condenser cycle Refrigeration cycle
Q
Generator Heat exchangerWork input rate
W
Equation 5:
𝐶𝑂𝑃 =𝑄̇𝑒 𝑊̇𝑐 Whereas:
𝐶𝑂𝑃 = coefficient of performance 𝑄̇𝑒 = rate of heat transfer (W)
𝑊̇𝑐 = electrical power input (W)
As seen in Figure 15 and Figure 16, the purpose of a refrigeration cycle is to transfer heat from the water to the refrigerant fluid and vice versa. The same principle applies to the evaporator and condenser cycles. The one difference is that the heat is transferred between air and water through convection and/or evaporation.
The evaporator cycle uses a BAC to transfer heat from the ambient air to cold evaporator water. This is due to the temperature of the evaporator water being lower than the temperature of the ambient air. The result is cool dehumidified air at an average wet-bulb temperature of 7°C that is used for underground ventilation. A BAC will either be a forced draft or an induced draft BAC with fans at the air inlet or air outlet respectively. It is not necessary for a BAC to be vertical as horizontal BACs can be used if adequate space is available (Amsted Industries, 2011). Figure 17 shows a typical vertical BAC used on a mine refrigeration and ventilation system (McPherson, 1993).
The condenser cycle uses cooling towers to transfer heat from the hot condenser water to the ambient air. With the temperature of the ambient air being lower than that of the condenser water sensible heat transfer occurs, also known as convection. Not only does sensible heat transfer occur but latent heat transfer, also known as evaporation, occurs when the water changes from phases. A condenser cooling tower usually utilises induced draft with fans situated on top of the tower as shown in Figure 17 (Amsted Industries, 2011; McPherson, 1993).
Figure 17: Typical cooling tower and vertical bulk air cooler (Amsted Industries, 2011; McPherson, 1993).
The numbered items below summarise the evaporator cycle of a typical refrigeration and ventilation system of mine.
1. The hot dam water is cooled to an average temperature of 30°C by a pre-cooling tower before it enters the evaporator cycle.
2. The water collected in the BAC sump enters the evaporator where the heat is transferred to the refrigerant fluid.
3. The water will now either be sent underground as service water or it continues to the BAC after being cooled to an average temperature of 2 °C by the evaporator.
4. Within the BAC, heat from the ambient air will be transferred to the cold evaporator water resulting in cool dehumidified ventilation air.
5. Large extraction fans at the return air way shaft draw the cool air from the BAC through the main shaft and mining levels.
6. The service water used through the mine is pumped to the surface hot dam after being utilised underground.
2.3.2 Refrigeration and ventilation system alterations and improvements
In general, a refrigeration and ventilation system is developed to produce either both cold service water and/or cool dehumidified ventilation air. Both these elements are essential for an underground mine to adhere to all the health and safety regulations. Unfortunately underground mine refrigeration and ventilation systems are operated inefficiently to adhere to these regulations. This presented
Cooled ventilation air Chilled water in Water return Ambient air in Packing Hot water in Cooled water out Ambient air in Packing Water droplets Heated air Sprays Fan
energy demand. This section will summarise various strategies that have already been implemented and proven successful.
Load management
Load management refers to the efficient management of the cooling load to reduce the electrical costs. However, load management does not reduce the amount of electrical energy utilised during the day, but only the electrical costs. This is achieved by spreading the load evenly during the Eskom off-peak hours to reduce the load in the peak hours. The load reduction entails switching off the entire refrigeration plant if and only if the necessary preparations have been achieved. Figure 18 shows the normal electrical power and the proposed load management profile for Mine A.
Figure 18: Load management power profile for Mine A.
The potential for a load management project can be identified by evaluating the refrigeration machine utilisation and thermal storage capacity. These two approaches have been developed and refined through numerous studies. The refrigeration machine utilisation is the percentage of electrical load used of the total installed capacity of the plant (van der Bijl, 2007). Whereas, thermal storage capacity refers the capacity to store sufficient cold water for underground usage during the Eskom peak hours (van der Zee, 2013).
A load management project will focus on using existing infrastructure to its maximum constraints during off-peak periods. This refers to the filling of chilled water dams to their maximum levels in order to supply underground usage during peak hours. The chilled water temperature should also be decreased to maximise thermal storage. An additional back-pass of chilled water to the hot dam or refrigeration machine will result in a decrease in compressor power consumption (ASHRAE, 1999; Calitz, 2006). 0 1500 3000 4500 6000 7500 9000 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 P ow er [ kW] Time [Hour]
Refrigeration and ventilation supply control
The operating conditions for a refrigeration and ventilation system on a mine are similar to that of a compressed air system. Cold service water and ventilation air are oversupplied to ensure that there are no short comings (Du Plessis, 2013). However, with current technologies, such as variable speed drives (VSDs), the supply can be reduced in accordance with the underground demand (Yu & Chan, 2010).
Some underground mines implement primitive electrical energy saving strategies on the refrigeration machines by reducing the evaporator and condenser water flow. Unfortunately this is achieved by throttling isolation valves resulting in inefficient pump operation. Results published by van der Zee (2013) depicted a 45% reduction in evaporator pump power with the use of a VSD rather than throttling an isolation valve. Figure 19 shows the normal electrical power and the proposed flow control profile for Mine A.
Figure 19: Flow control power profile for Mine A.
VSDs allow electrical motors to be operated at variable speeds and rotational forces. This is achieved by using a variation in output voltage, current and frequency, also known as pulse width modulation. Using VSD on electrical motors not only reduces the power consumption but increases the energy efficiency, power factor and operational life (Saidur et al., 2012). Figure 20 illustrates the correlation between the potential electrical energy savings and average electrical motor speed reduction (Saidur, 2010). 0 1500 3000 4500 6000 7500 9000 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 P ow er [ kW] Time [Hour]
Figure 20: Potential electrical energy savings and electrical motor speed.
Various studies performed by Du Plessis (2013) show that a variable flow strategy has no or a negligibly small effect on the service delivery of a mine. In contrast a variable flow strategy can be accountable for electrical savings of up to 60% with only a 30% reduction in flow (Berkeley Law, 2008). The success of a variable flow strategy is directly dependent on the layout and operation of a refrigeration and ventilation system. Therefore each auxiliary component has been thoroughly discussed in Section 2.3.1. An example of the refrigeration and ventilation operational specifications of Mine C is depicted in Table 2 (Du Plessis, 2013).
Table 2: Specifications of the refrigeration and ventilation system on Mine C (Du Plessis, 2013).
Refrigeration and ventilation system
Air inlet wet-bulb temperature [°C] 22
Water temperature from underground [°C] 30
Flow rate to underground [ℓ/s] 313
Chilled water dam temperature [°C] 6
Plant COP 5.00
Plant cooling capacity [kW] 42000
Barometric pressure [kPa] 90
Refrigeration machines
Machine type York
Cooling capacity per machine [kW] 13300
Evaporator outlet temperature [°C] 5.9
Condenser inlet temperature [°C] 18.5
Evaporator water flow rate [ℓ/s] 300
Condenser water flow rate per machine [ℓ/s] 600
Machine COP 6.65 Refrigerant R134a 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 P ot ent ial ener gy savi ngs [% ] Speed reduction [%]
Water pumps
Number of evaporator pumps 4
Evaporator pump motor rating [kW] 90
Number of condenser pumps 5
Condenser pump motor rating [kW] 185
BAC return water pumps
Number of return pumps 3
Return pump motor rating [kW] 75
Pre-cooling towers
Number of pre-cooling towers 8
Water inlet temperature [°C] 28
Water outlet temperature [°C] 24
Condenser cooling towers
Number of condenser cooling towers 4
Water inlet temperature [°C] 32
Water outlet temperature [°C] 27.5
Bulk air coolers
Number of bulk air coolers 3
Water inlet temperature [°C] 3
Water outlet temperature [°C] 14
Air inlet wet-bulb temperature [°C] 22
Evaporator water flow control
Electrical energy savings on the evaporator cycle can be achieved with two different variable flow strategies. The first strategy is by reducing the water flow, to an acceptable minimum, through the evaporator. This will result in a reduced compressor and evaporator pump power consumption. The reduced compressor power consumption is due to the guide vanes controlling at a lower cooling load (Du Plessis et al., 2013). The variable evaporator water flow control can also incorporate a more effective level control of the chilled water dam.
The second strategy on the evaporator cycle is to vary the cold water flow through the BAC to maintain a safe cooling air temperature. With the sessional change in ambient wet-bulb temperature, the amount of cooling needed during the winter will automatically be less. A reduction in compressors and pump power can be achieved by reducing the flow through the BAC according to the desired wet-bulb temperature.
